How To Analyse Public Policies: The Basics
There is a lot of discussion about which policies are good, or bad, and how to measure them. This first article covers the following perspectives on public policy decision making, through perspectives Pareto, Edgeworth, Kaldor-Hicks, and Okun. We consider efficiency, equality, ‘goodness’, or ‘badness’. We shall later cover specific perspectives, such as Utilitarian, Rawl, and Nozick and Smith.
This article will contain a lot of graphs; I cannot recommend Bellinger’s “The Economic Analysis of Public Policy” enough if you want to read further. However, this will provide a good basic grounding in how to analyse and consider public policies, and whether to implement them. At least, I hope it helps you consider how we should measure whether a policy is helpful, who it helps, and is it good to implement.
The Foundation of Public Policy:
When you begin to consider public policy, we have to consider the basic views of whether a policy is good, and whether we should implement it, and what are the side effects, and who are the stakeholders affected. We can begin with what is considered the fundamental rule of public policy:
The Fundamental Rule of Policy Analysis
Adopt a policy if the total benefits are greater than the total costs.
Seems simple, right? But what are the benefits? What are the costs? How do we define greater or worse? If I take $10 from one man, and keeps $4 for myself, and give two other people $3, is this good? I mean, 3 people have benefitted while only one has suffered.
Is it a good policy if I tell you that the victim in this case is rich? How rich does he have to be before you agree it’s okay to take him money to give to other people? Rawl would say that as long
Or does the simple harm of one to benefit others make this a bad policy? Pareto would say that unless the loser were compensated fully (by giving him equivalent of $10), it could not be. Nozick would say it could never be; the loser must choose to do so voluntarily.
So in this case, we must divide our measurements of ‘goodness’ further. There are three factors we can split most policies along to measure them (Bellinger, 2007).
Efficiency: Maximising the greatest welfare for society given scarce resources. Making the overall pie bigger.
Equity: Income equality, freedom, minimum standards of treatment, equal rights, equal opportunity. Making sure everyone gets a more equal slice of the pie.
Political Practicality: Political, popular support, and legal standing. Making sure that it is legal, and politically possible to touch the pie at all.
The first two are usual economic concerns; the third a practical concern, especially in a democracy. Even if you have the perfect tax plan, you need to get the support of a majority of the legislature and the executive to agree. It is not enough to be right; you need to ‘sound right’, and depending on the country, require the legal system, Constitution, and history to agree. If you’re the leader of a dictatorship, you’d still need to make sure to not generate sufficient popular unrest (revolt) or noble unrest (a coup) to lose your position.
Other language and concerns often alluded to are (Bellinger, 2007):
Rights: A power or privilege that one is entitled to. A level of minimum standards that are acceptable. E.g. The Human Rights Act (U.K.); the U.S. Constitution; the French Constitution.
Freedom: A lack of constraints or limits on an individuals’ actions. E.g. freedom of speech, freedom of opportunity, freedom of association.
Equality: How to distribute the rights, incomes, legal status, and other aspects of society.
Utility: A measurement from the welfare, or how much good, a person receives. E.g. If you have a cheeseburger, and you like it, you have more utility (happiness/welfare) than you did before you had a cheeseburger.
Generally speaking, Libertarians/traditional Liberals consider upholding the rights and freedoms of the individual above that of the collective society; Socialists believe that governments have the right to place the needs of the many above the needs of the few, and Communists believe that only the needs of the many matter. There are many ways to measure how good a policy is, and whether the good outweighs the bad. We shall cover a selection of perspectives in this, and the next, article.
Pareto, an Italian economist, most famous for this Pareto distribution (that 80% of land is owned by 20% of people, and further shown to exist in terms of many social situations; such as dating and wealth inequality) also provided rules for assessing whether a public policy should be implemented.
Pareto Optimality: A situation is Pareto optimal if it is impossible to make any person better off without making at least one person worse off. All possible improvements of mutually beneficial trade must be done.
Pareto Improvement: If something can improve someone without making someone else worse off. Can also be achieved if the person who losses receives compensation equal to their loss. Losers must be compensated.
So if there is a redistributive policy where Person X loses $10 of income, but Person Y gains $20 of income benefits, then it is not a Pareto improvement until Person Y compensates the loser, Person X, with worth of $10.
So what is Pareto optimal? If you have $1, and you want a can of coke that costs $0.50 from the producer, and less than that to produce, you can buy that (by your consent). You gain the coke that you believe is worth more than having that $1; the seller gains $0.50 in profit (minus taxes), the producer gains $0.50 from the seller. Everyone has benefitted from this action; everyone believes that they gained from this action. It is Pareto Optimal, and if it can be done, and the people want it to be done, then it must be done to improve society.
Across a society? It’s more difficult to measure. For example, a government to fund education. People, even people who do not want to use this, have to pay for this via taxes. For this to be a Pareto improvement, everyone who pays taxes would have to benefit from this. There are multiple ways to argue this:
Having an educated populace means less crime
Having an educated populace means a richer populace, which creates demand for goods (that pay you) and pay taxes (that pay for other services you use).
Having an educated populace (hopefully) means a smarter populace; less likely to want democratic or political decisions that will negatively impact your life
Whether or not you agree, there is a general rate of return on investment in education for society, ranging with a return of 8%-9% for society (Psacharopoulos and Patrinos, 2018). You could argue then it is likely Pareto optimal, but not everyone would agree.
How can we work out where a trade benefits both parties? Purely theoretically, we could construct a box where we measure the utility curves of two people with goods, and where they would be able to trade and gain more pleasure from the trade than lost via trade. One visualisation is the above Edgeworth Box.
An Edgeworth Box (Bellinger, 2007) begins at the endowment point (the goods people begin with) and assumes no further production of goods. We now try to find the balance between Schwartz and Dottie. We compare each parties indifference curves, until they can trade and be improved. The area between the two curves, the core of the areas, is the area that can be improved. When you trade here, both of them move beyond their own indifference curves without loss.
It is a visualisation of the benefits of trade in individual utility, but it works with current existing micro-economic theory on maximising utility individually, and combines it into a trade model with two parties.
A policy should be adopted if the winners could in principle compensate the losers. It requires that total benefits outweigh total costs, but does not require repayment to the losers. Often measured in monetary value.
In the above educational example, we could simply state that investment on education always has returns on investment for society, and worry a lot less about how to justify it to the losers. The problem with this is, while it is convenient to simply ignore the issue of repayment, it does run roughshod over the rights of people to generally enjoy their own goods and property in peace in the name of societal improvement.
This should be balanced with Libertarian wariness; how many rights can we ignore? How much can we take for the greater good? Even if you have a perfect plan, it doesn’t give you the right to simply cast your hand and redistribute wealth, money, and assets as you see it; people have the right to refuse you, or at least be compensated for their loss.
However, sometimes it is for the greater good; policies that improve innovation will improve an economy’s long-term growth even at the cost of the jobs of the lower class, as well as bringing over immigrants to counter the massive pension scheme most governments have. In some countries, they believe that the security of the country must take precedence over the economy of that country.
This is more of a political and philosophical issue than a measurement, but provides a more lenient view than Pareto for politicians or analysts who need to get things done without worrying about breaking eggs.
Okun’s Leaky Bucket
Shocking to no one who pays attention to governments, all government policies ‘leak’ money (measured when Tax Received for Redistribution > Income Received by Poor). His theory focuses first on identifying and fixing those leaks, rather than measuring them, and designing policy to reduce this leaking. Any policy that completely ‘leaks’ cannot be justified.
Potential Sources of Leaks (Bellinger, 2007):
Diversion of Funds to the Non-Poor
Disincentives to Work
Disincentives to Save and Invest
Socio-economic leaking: effects of transfer payments on work ethic or such among the lower classes.
A lot of newer policies are designed to reduce this: for example, a UBI would be a flat payment given to a citizen; you don’t need to pay for someone to send it, you don’t need anything beyond a simple birth certificate/passport, and once set up, runs by itself. No one needs a salary, money can’t go missing, simple.
The same with Milton Friedman’s Negative Income Tax (1968); set a simple rate, and when you submit it to the IRS (or your local tax authority), get given a payment back. It can be set up to be automatic, and require absolutely nothing except a value entered into a website. Simple, no salary, efficient.
One problem with a lot of means-tested benefits requires tests, checks, visits, research, doctor’s notes, and other things that all reduce the amount of money that goes to the person, rather than trying to prove they don’t need the payment at the beginning. However, in a 3-year Negative Income Tax experiment with 1,300 families in Pennsylvania in 1967 found that the administrative costs were actually over half of the total costs including payments (Rossi and Lyall, 1976).
We have covered the basics of public policy; what are the terms used, what are the aspects of public policy that must be considered, and some of the definitions for whether a public policy is a good or bad one, whether you measure it by net benefit, Pareto optimality, or Kaldor-Hicks’ principle. I hope this has been a good foundation in this analysis.
A good public policy must be well-reasoned, and well-argued. Choosing what sounds nice over what is good may be populist; it may even get you elected. But will it lead to long-term success for your society? Will it improve the lives of those that the government must care for? Or put ideology over the comfort and welfare of the people?
In the next article, we shall cover three perspectives in public policy; Utilitarian, Rawlsian, and Nozickian, and provide formulas, graphs, and definitions to understand the practical use of these perspectives in creating and evaluating public policy.
If you’ve enjoyed reading this, please consider following me on Twitter @LeonDeclis or on Apple News on the Idea Meritocracy channel, or a Facebook page at @IdeaMeritocracyEcon. There is also an RSS feed option on this website for direct updates! Have a nice day!
If you’d like an article on public policy in China, please consider this article about monetary policy and Chinese customers.
If you’d like an article on immigration public policy in the United Kingdom, please consider this article about employment, wages, and immigration.
In all articles, I provide as much information for sources as possible, including links. I encourage everyone reading this article to read deeper, and make their own conclusions. For students, links are here so they can read the original source themselves. Most sources are linked the first or second time they appear in the article.
Bellinger, W., (2007), “The Economic Analysis of Public Policy”, published by Routledge Taylor and Francis, 1st Edition
Friedman, M., (1968), "Firing Line with William F. Buckley Jr.: The Economic Crisis”, published on Firing Line with William F. Buckley, retrieved from https://www.youtube.com/watch?v=Cjj-fCKGdts on 15th September 2019.
Psacharopoulos, G. and Patrinos, H.A., (2018), “Returns to investment in education: a decennial review of the global literature”, published by Education Economics, 26:5, retrieved from http://datatopics.worldbank.org/education/files/GlobalAchievement/ReturnsInteractive.pdf on 15th September 2019
Rossi, P., and Lyall, K., (1976), “Reforming Public Welfare: A Critique of the Negative Income Tax Experiment”, published by the Russell Sage Foundation, New York, U.S.